ms_thesis_hli_4_4_2008
TRANSCRIPT
-
7/29/2019 MS_Thesis_HLi_4_4_2008
1/100
r
SELECTIVE CATALYTIC OXIDATION OF HYDROGEN SULFIDE FROM SYNGAS
by
Huixing Li
B.S., South China University of Technology, 2003
M.S., Universit de Savoie, 2006
Submitted to the Graduate Faculty of
Swanson School of Engineering in partial fulfillment
of the requirements for the degree of
Master of Science
University of Pittsburgh
2008
-
7/29/2019 MS_Thesis_HLi_4_4_2008
2/100
UNIVERSITY OF PITTSBURGH
SWANSON SCHOOL OF ENGINEERING
This thesis was presented
by
Huixing Li
It was defended on
April 1st, 2008
and approved by
Jason D. Monnell, Research Assistant Professor, Department of Civil & Environmental
Engineering
Leonard W. Casson, Associate Professor, Department of Civil & Environmental Engineering
Thesis Advisor: Radisav D. Vidic, Professor and Chair, Department of Civil & Environmental
Engineering
ii
-
7/29/2019 MS_Thesis_HLi_4_4_2008
3/100
Copyright by Huixing Li
2008
iii
-
7/29/2019 MS_Thesis_HLi_4_4_2008
4/100
SELECTIVE CATALYTIC OXIDATION OF HYDROGEN SULFIDE FROM
SYNGAS
Huixing Li, M.S.
University of Pittsburgh, 2008
In order to obtain a non-corrosive fuel gas (syngas), which is derived from coal by Integrated
Gasification Combined Cycle (IGCC) technology and used in power plants, hydrogen sulfide
(H2S), which is generated during the gasification process due to sulfur contained in coal, should
be removed to protect instruments, especially turbines, from corrosion. To improve H2S removal
efficiency and develop excellent regenerative catalyst, we conducted the following research.
Simulated syngas was introduced into a fixed-bed quartz reactor where carbonaceous sorbents
(which are excellent sorbents and have lower price) were positioned to capture H2S. Tail gases
from the outlet of the reactor including H2S, COS, and SO2 were continuously monitored by a
residual gas analyzer (or mass spectrometer) to determine the capacity of H2S uptake and
selectivity of adsorption/oxidation by different sorbents.
Carbonaceous materials including carbon black, graphite and activated carbon fibers
(ACFs) were compared for the application in desulfurization. Rare earth metal oxides (La2O3 and
CeO2) were investigated and used to modify ACFs due to their potential to effectively remove
H2S and multicycle regenerative ability. Water vapor and temperature effects on H2S removal
were studied.
Functional groups on carbonaceous materials were determined and the mechanism of the
promotion of H2S uptake by basic functional groups was proposed. Through the determination of
activation energy of desorption of sulfur species from sulfided sorbents, it is concluded that
iv
-
7/29/2019 MS_Thesis_HLi_4_4_2008
5/100
chemisorption is the dominant mechanism at higher sulfurization temperature, while
physisorption is the controlling process at lower temperature. At the temperature ranging from
110 to 170 C, the best H2S-uptake capacity was obtained because chemisorption and
physisorption are both present and water film on the surface of sorbents is ideally maintained.
The observation of sulfurization and regeneration of sorbents and by-products related that
nitrogen could only remove physically adsorbed H2S and hydroxide could, to some extent,
restrain the formation of by-products (the reaction between COS or SO2 and hydroxide). ACFs
modified by metal compounds showed excellent H2S-uptake capacity (up to 35 mg H2S/g
Sorbent) in the 1
st
cycle but the capacity in the subsequent cycles were much lower than the 1
st
cycle because regeneration gas (nitrogen) could not recover the chemically adsorbed sulfur
species.
Key words: Hydrogen Sulfide, Carbon, Syngas, Sulfurization, Adsorption, Regeneration
v
-
7/29/2019 MS_Thesis_HLi_4_4_2008
6/100
TABLE OF CONTENTS
TABLE OF CONTENTS ........................................................................................................... VI
LIST OF TABLES ...................................................................................................................... IX
LIST OF FIGURES..................................................................................................................... X
ACKLOWDEGMENT ............................................................................................................. XII
1.0 INTRODUCTION........................................................................................................ 1
2.0 LITERATURE REVIEW............................................................................................ 5
2.1 HYDROGEN SULFIDE SOURCES.................................................................. 5
2.2 TECHNOLOGIES FOR H2S REMOVAL........................................................ 7
2.2.1 Use of Carbonaceous Based Materials to Remove H2S................................. 9
2.2.2 The Usage of Metal Oxide to Remove H2S................................................... 13
2.2.3 The Usage of Impregnated Sorbents to Remove H2S.................................. 17
2.2.4 Other Methods to Remove H2S ..................................................................... 20
2.3 EFFECT OF PROCESS CONDITIONS ON H2S REMOVAL .................... 24
2.3.1 Water Vapor Effects....................................................................................... 24
2.3.2 Temperature and Surface Area Effect ......................................................... 26
2.3.3 Impact of Surface Functional Groups .......................................................... 28
2.4 MECHANISM OF H2S REMOVAL BY CATALYTIC REACTIONS ....... 29
2.5 SUMMARY........................................................................................................ 34
vi
-
7/29/2019 MS_Thesis_HLi_4_4_2008
7/100
3.0 MATERIALS AND METHODS .............................................................................. 36
3.1 MATERIALS..................................................................................................... 36
3.1.1 Materials.......................................................................................................... 36
3.1.2 Simulated Syngas............................................................................................ 37
3.2 EXPERIMENTAL SETUP AND SAMPLE PREPARATION..................... 38
3.2.1 Experimental Setup ........................................................................................ 38
3.2.2 Impregnation of Activated Carbon Fibers with Lanthanum Compounds 40
3.3 SURFACE FUNCTIONALITIES ON CARBONACEOUS MATERIALS. 40
3.4 SULFURIZATION AND REGENERATION EXPERIMENT .................... 41
3.5 THERMOGRAVIMETRIC ANALYSIS ........................................................ 42
4.0 RESULTS AND DISCUSSION ................................................................................ 43
4.1 H2S UPTAKE ONTO CARBONACEOUS SORBENTS............................... 43
4.1.1 Functional Groups on Carbon Surface ........................................................ 43
4.1.2 H2S Uptake onto Carbonaceous Materials................................................... 45
4.2 EFFECT OF SURFACE FUNCTIONALITIES ON H2S UPTAKE ............ 48
4.3 EFFECT OF WATER VAPOR ON H2S UPTAKE ....................................... 50
4.4 TEMPERATURE EFFECT ON H2S UPTAKE............................................. 52
4.4.1 Effect of Temperature on H2S uptake .......................................................... 52
4.4.2 Effect of Heat Pretreatment at Different Temperature.............................. 55
4.5 REGENERATION OF SORBENTS................................................................ 58
4.5.1 Sulfurization and Regeneration of Lanthanum Oxide and Cerium Oxide58
4.5.2 Sulfurization and Regeneration of ACF10................................................... 62
4.5.3 Sulfurization and Regeneration of Lanthanum-Modified ACFs ............... 64
vii
-
7/29/2019 MS_Thesis_HLi_4_4_2008
8/100
4.5.4 Sulfur Byproducts........................................................................................... 70
5.0 SUMMARY AND CONCLUSIONS ........................................................................ 72
6.0 FUTURE WORK ....................................................................................................... 74
APPENDIX.................................................................................................................................. 75
BIBLIOGRAPHY....................................................................................................................... 83
viii
-
7/29/2019 MS_Thesis_HLi_4_4_2008
9/100
LIST OF TABLES
Table 1. Typical gas compositions (% by volume)......................................................................... 7
Table 2. The application of metal oxide in desulfurization .......................................................... 15
Table 3.Composition of simulated syngas under dry and wet condition..................................... 38
Table 4. Functionalities on the surface of carbon blacks, graphite and ACFs.............................. 44
Table 5. Surface area of carbon samples and their H2S-uptake capabilities................................. 46
Table 6. Breakthrough time and capacity of ACF10 under wet condition. .................................. 51
Table 7. Effect of Pretreatment Temperature on the Capacity of H2S Uptake............................. 56
Table 8. Comparison of the results between our experiment and reference................................. 58
Table 9. Experiment results of sulfurization and regeneration of various sorbents...................... 62
Table 10. Functionalities on the surface of carbon blacks and graphite....................................... 76
ix
-
7/29/2019 MS_Thesis_HLi_4_4_2008
10/100
LIST OF FIGURES
Figure 1. A schematic flow diagram of an IGCC plant. ................................................................. 2
Figure 2. Schematic process flow diagram of a basic Claus process unit. ..................................... 8
Figure 3. Scanning electron microscopy picture of an activated carbon cloth ............................. 12
Figure 4. Schematic of movement between impregnated solution and sorbents .......................... 19
Figure 5. Flow diagram for a phototrophic microbial acid gas bioconversion plant.................... 21
Figure 6. Gas-lift reactor where H2S is injected at the bottom of the riser................................... 23
Figure 7. Sulfurization of ZnO with H2S...................................................................................... 32
Figure 8. Ion migration and sulfidation penetration ..................................................................... 33
Figure 9. Particle size and distribution of La2O3. ......................................................................... 37
Figure 10. Schematic of carbon sorbents breakthrough capacity test........................................... 39
Figure 11. Basic and acid functional groups on the surface of each carbon material................... 44
Figure 12. Breakthrough curves of carbon black samples and graphite....................................... 46
Figure 13. Breakthrough curves of ACFs..................................................................................... 47
Figure 14. Relationship of capacity of H2S uptake and basicity functionalities........................... 49
Figure 15. Possible basic surface sites on carbon surface having -pyrone-like structure. .......... 49
Figure 16. Interreaction between H2S and basic sites on carbons surface.................................... 50
Figure 17. Sulfurization under wet condition at 25, 110, 170 and 210C respectively. .............. 51
Figure 18. Activation energy of desorption analysis employing TGA......................................... 54
x
-
7/29/2019 MS_Thesis_HLi_4_4_2008
11/100
Figure 19. Arrhenius plot lnk vs. 1/T for the activation energy of desorption ............................. 55
Figure 20. ACF10 heated at 500 C and 300 C for 4 hours and sulfurized at room temperature57
Figure 21. Sulfurization (S) and regeneration (R) of 6 g CeO2 for 3 cycles................................. 60
Figure 22. Sulfurization (S) and regeneration (R) of 6 g La2O3 for 3 cycles ............................... 61
Figure 23. Sulfurization (S) and regeneration (R) of 2.0 g ACF10 under wet condition. ............ 63
Figure 24. Sulfurization (S) and regeneration (R) of ACF-LaCl3 ................................................ 65
Figure 25. Sulfurization (S) and regeneration (R) of ACF-La2O3-NaOH. ................................... 67
Figure 26. Sulfurization and regeneration of ACF- LaCl3-NaOH................................................ 68
Figure 27. Particle size and distribution of ACF10. ..................................................................... 75
Figure 28. Particle size and distribution of CeO2. ........................................................................ 76
Figure 29. Sulfurization of ACF 10 with simulated syngas whose ratio of H2S and O2 is 1 : 2. . 77
Figure 30. Sulfurization of dry mixed sorbents of ACF and CeO2............................................... 77
Figure 31. TGA run of ACF10 without sulfurization and after sulfurization............................... 78
Figure 32. TGA kinetic analysis of ACF10 sulfurized at 25 C................................................... 79
Figure 33. TGA kinetic analysis of ACF10 sulfurized at 110 C................................................. 80
Figure 34. TGA kinetic analysis of ACF10 sulfurized at 210 C................................................. 81
Figure 35. Elemental sulfur desorbed from ACF accumulated at the outlet of the tube of TGA. 82
xi
-
7/29/2019 MS_Thesis_HLi_4_4_2008
12/100
ACKLOWDEGMENT
I am grateful to my advisor Dr. Radisav D. Vidic for his teaching, advice and support in my
research work. His precise research attitude, insight into different research field, social
acceptability and wise leadership deeply impress me. I cannot be thankful enough to my research
co-advisor Dr. Jason D. Monnell for participating in my research experiments, inspiring me,
offering a variety of brilliant suggestions and solving a lot of problems.
I am thankful to Dr. Leonard W. Casson and Dr. Ronald Neufeld for their teaching. This
study is supported by the National Energy Technology Laboratory (NETL), USDOE. I also wish
to thank Maryanne Alvin from NETL, Esteban Broitman and Jim Miller from Carnegie Mellon
University.
I would like to thank my colleagues and friends, Xihua Chen, Ravi Bhardwaj, Wei Sun
and Kent Pu for their support and help on daily work and study. Last but not the least, great
appreciation goes to my parents who show love and support to me.
xii
-
7/29/2019 MS_Thesis_HLi_4_4_2008
13/100
1.0 INTRODUCTION
Hydrogen sulfide is an undesirable byproduct of using fossil fuels for energy since it is corrosive
to process equipments. While the Clause process is able to remove up to 99% of H2S, for fuel
cell and syngas purpose, much more H2S need to be removed. The presence of sulfur is a
persistent problem related to fossil fuels and coal, and it must be removed before the fuels can be
used as energy sources or chemical feedstock. The main usage of coal lies in generating
electrical power and more than half of the total electricity in the U.S. is generated from coal. As
our reliance on fossil fuels continues, coal will play a more and more important role because of
its lower price [1]. Hydrogen sulfide (H2S) exists naturally in many natural gas wells, and is
produced largely in the desulfurization of petroleum stocks. For example, there are a few parts
per million (ppm) to hundreds of ppm sulfur in gasoline or diesel fuels, up to 20 ppm sulfur in
pipeline natural gas, several hundred ppm to more than 1.0% in coal-derived gaseous fuels
(depending on the type of coal, the sulfur content in the feedstock, and the process). Though H2S
has a considerably high heating value, its use as a fuel is impossible because one of its
combustion products is SO2 which is harmful to environment due to the possibility to lead to acid
rain. Further, H2S is corrosive to power equipment and there is also sulfur deposition as well.
When these fuels are used to produce hydrogen for proton exchange membrane (PEM) fuel cells,
the sulfur content will severely poison fuel processing catalysts and platinum electrocatalysts.
1
-
7/29/2019 MS_Thesis_HLi_4_4_2008
14/100
Therefore, it is important and necessary to remove H2S from these streams to allow further
processing [2,3].
The integrated gasification combined cycle (IGCC) is considered to be the most
efficient and environmentally acceptable technology for power generation from coal (Figure 1)
[4]. The removal of pollutants from coal-derived fuel gas is needed for effectively using this
technology. Among the pollutants, H2S which is formed from sulfur contained in coal must be
removed from the hot fuel gas not only to protect equipments against corrosion in the later stages
of the process, but also to meet the environmental legislation for sulfur emissions [5].
Figure 1. A schematic flow diagram of an IGCC plant. The gasification process can produce
syngas from high-sulfur coal, heavy petroleum residues and biomass in a gasification
unit. The gasification process produces heat, and this is reclaimed by steam "waste heat
boilers". Steam turbines use this steam [4].
Syngas is a gas with abundance of CO and H2, which are major components to
generate electric power. Sulfur contained in the coal is transformed into H2S and becomes part of
2
-
7/29/2019 MS_Thesis_HLi_4_4_2008
15/100
the syngas during coal gasification process. Therefore, H2S needs to be removed from syngas
before it is used to generate electric power because it is malodorous and corrosive.
There are many commercial treatment techniques that are used to remove H2S
including adsorption by activated carbon, condensation, chemical oxidation, incineration or
catalytic combustion and dry/wet absorption. The disadvantage of these techniques is that hot
coal gas (Generally, the temperature of syngas produced from coal gasification reaches 500 C or
above [6].) must be cooled down to ambient temperature or around 150 C for desulfurization
and then preheated to high temperature before being fed into gas turbine. In this way the thermal
efficiency of the system is significantly reduced. It is for this reason that high temperature
desulfurization technique has attracted more and more attention due to the fact that it can reduce
H2S down to 10 ppm level and avoid heat loss [6].
Porous materials show different activity and selectivity towards H2S adsorption
and/or oxidation. Zeolites show good activity at 300 C, but they do not have good selectivity.
Alumina and silica with macropores show both poor activity and selectivity. Metal oxides,
especially V2O5, TiO2, Fe2O3, CuO and ZnO show high activity for H2S oxidation and
regeneration capability [3].
Impregnated activated carbons have been used to remove H2S efficiently. Alkaline
materials such as NaOH, Na2CO3, KOH and K2CO3 were impregnated into activated carbons. It
was found that the removal efficiency with these impregnated sorbents ranks as NaOH > Na2CO3
> KOH > K2CO3 [7].
Te regeneration of the spent sorbents remains a problem, since the material can
degrade and lose its functional chemistry. For example, the thermal desorption of sulfur species
from sulfided activated carbon, the formation of CO or CO2 leads to the change of surface
3
-
7/29/2019 MS_Thesis_HLi_4_4_2008
16/100
characteristics of the initial carbon sorbent. On the other hand, the sulfur species are poorly
removed by rinsing the activated carbons with water, because large amounts of sulfur species can
remain trapped in pores due to the poor penetration of solvent [8].Regeneration of spent carbon
sorbents by thermal treatment with inert gas or gas mixture containing oxygen and/or water
vapor is capable of partial regeneration of its initial capacity. This is due to the fact that
elemental sulfur or sulfuric acid is strongly bound to active sites [9].
Given the limitation of current materials, it is necessary to develop H2S-uptake
sorbents with high removal efficiency, high catalytic activity, high selectivity for desired
products and good regeneration ability. In this thesis, adsorption, oxidation and catalytic property
of carbonaceous sorbents, impact of functional groups on carbon surface, effect of sulfurization
temperature, water vapor and by-products, performance of rare earth metal modified sorbents
and regeneration during H2S removal process was investigated.
4
-
7/29/2019 MS_Thesis_HLi_4_4_2008
17/100
2.0 LITERATURE REVIEW
2.1 HYDROGEN SULFIDE SOURCES
H2S is an undesired product of many industrial processes and it is primarily formed in nature by
decomposition of organic materials by bacteria while it is also a constituent of natural gas, sulfur
deposits, volcanic gases and sulfur springs [10]. Hydrogen sulfide (H2S) is a toxic gas that can
lead to personal distress even at a low concentration while at a higher concentration it can result
in loss of consciousness, permanent brain damage or even death due to the neurotoxic effect [10].
In the Earths past H2S came from volcanic eruptions, which emitted CO2 and
methane into the atmosphere causing global warming. Under elevated temperatures, the oceans
were warmed resulting in lower dissolved oxygen, which would otherwise oxidize H2S [11]. It is
believed that volcanoes make the largest contribution of H2S to the atmosphere from active and
inactive volcanoes, as well as lava flows, domes, fumaroles, erupting lava and submarine
hydrothermal areas. The major gas compositions include H2O, SO2 CO2, and HCl, with minor
components being N2, H2, CO, H2S, HF and HBr. The gas composition is highly variable in
terms of different locations and depends on the individual magma geochemistry and nature of an
eruption [12].
H2S exists in the biogeochemical cycle as the product of anaerobic oxidation by
sulfur-reducing bacteria [11]. Anaerobic bacteria use sulfate as the electron acceptor in the
5
-
7/29/2019 MS_Thesis_HLi_4_4_2008
18/100
respiration process, resulting in the production of H2S. They exist at a wide range of pH, pressure,
temperature, and salinity conditions and are widely distributed in habitats. In anaerobic organic
rich soils, H2S is commonly present due to the reducing nature of the sediments and the
precursor SO42
is the second most common anion in seawater [12]. Bacteria break down organic
matter produce H2S. These microorganisms favor low-oxygen environments, such as in swamps
and sewage.
Small amounts of H2S exist in crude petroleum while up to 90% in natural gas. About
10% of total global emissions of H2S are due to human activity including petroleum refineries,
coke ovens, paper mills (using the sulfate method), and tanneries. Normal concentration of H2S
in clean air is about 100-200 ppb [11]. Burstyn et al. [13] conducted a survey of average
concentrations of SO2 and H2S at rural locations in western Canada and determined that the
average concentration is 0.1-0.2 ppb for H2S, and 0.3-1.3 ppb for SO2. The concentrations vary
at different locations and are affected by seasons. Measured concentrations were affected by the
wells, especially those within 2 km of monitoring stations. Locations close to sour gas wells and
flares showed high H2S concentrations and it is determined that oil and gas extraction activities
contribute to air pollution in rural areas of western Canada.
IGCC plants produce large quantity of H2S in coal-derived fuel gas. The commercial
success of IGCC is greatly dependent on the accompanying air-pollution control systems. The
success of such plants is essential because the IGCC technology with an efficiency of 52% is
considered to be one of the key potential technologies that would meet the energy and
environmental demands [1]. Table 1 shows the components of fuel gas or simulated fuel gas
from different plants. From these values we determined a balance gas composition that was
replicated for this study.
6
-
7/29/2019 MS_Thesis_HLi_4_4_2008
19/100
Table 1. Typical syngas compositions (% by volume). Syngas in this research was simulated
according to the composition of syngas from NETL (the cooperator of this research).
Gas H2 CO CO2 H2O H2S N2 O2 Ref.
O-blown gasifier 27.7 39.4 13.1 18.4 1.1 0 0 [14]
Air-blown gasifier 14.2 23.1 5.8 6.6 0.5 49.8 0 [14]
Syngas stream 35 53 12 0 0 0 0 [15]
NETL Unshifted Gas 25-30 35-40 12-15 0.1-10 0.1-0.4 10-18 0.1-0.4 [16]
NETL Shifted Gas 45-50 0 34-38 0.07-7.4 0.28-0.3 11-12 0.28-0.3 [16]
This Thesis 24.4 34.6 15.2 0.4 0.4 16.8 0.4
2.2 TECHNOLOGIES FOR H2S REMOVAL
Many commercial technologies based on wet/dry adsorption and oxidation have been used for
H2S removal from gases. The Claus process, invented over 100 years ago, is the most significant
gas desulfurizing process for recovering elemental sulfur from gaseous hydrogen sulfide (Figure
2) [17]. Dry catalytic processes based on the selective catalytic oxidation of H2S to elemental
sulfur have been developed [2]. Two dry Claus processes were developed to avoid the
shortcoming of wet-Claus processes: Mobils direct-oxidation process developed by Mobil AG
Company in Germany and Super-Claus process developed by Comprimo Company in The
Netherlands [18]. Both utilize one-step recovery of elemental sulfur from Claus tail gas by the
following catalytic reaction:
H2S + 1/2 O2 1/n Sn + H2O (n = 6 ~ 8)
7
-
7/29/2019 MS_Thesis_HLi_4_4_2008
20/100
Figure 2. Schematic process flow diagram of a basic Claus process unit. The Claus technology
can be divided into two process steps, thermal and catalytic. In the thermal step, H2S-
laden gas reacts in a substoichiometric combustion at temperatures above 850 C such
that elemental sulfur precipitates in the downstream process gas cooler. In the catalytic
step the process continues with activated Al2O3 or TiO2, and serves to boost the sulfur
yield [17].
The catalyst used by Comprimos Super-Claus process is alumina-supported iron
oxide/chromium oxide, and the catalyst used by Mobils direct-oxidation process is a TiO2-based
catalyst. However, there are some adverse portions of these processes, such as toxic properties of
chromium oxide and the formation of SO2 byproducts by way of the following reactions.
H2S + 3/2 O2 SO2 + H2O
1/n Sn + O2 SO2
3/n Sn + 2 H2O 2 H2S + SO2
8
http://upload.wikimedia.org/wikipedia/en/8/8c/ClausPlt.JPGhttp://upload.wikimedia.org/wikipedia/en/8/8c/ClausPlt.JPG -
7/29/2019 MS_Thesis_HLi_4_4_2008
21/100
Many factors that affect the selection of the gas-treating process have been
investigated, including: the volume, temperature, and pressure of influent gas, space velocity, the
selectivity required, the desirability of sulfur recovery, the types and concentrations of impurities
present, surface characteristic of sorbents, the regeneration capability of sorbents, etc. [2].
The Claus process has high efficiency ranging from 94% to 97% and produce
relatively small ecological impact when coupled with a tail-gas treatment unit. Moreover, its
problems can be easily solved by using the correct stoichiometric amount of air, fuel gas, and
acid feed gas. The capabilities of the conversion process have been improved, including
innovations and optimization of process as well as reducing operating costs [2].
2.2.1 Use of Carbonaceous Based Materials to Remove H2S
All carbon blacks (CB) have chemisorbed oxygen complexes (i.e., carboxylic, lactonic, phenolic
groups, etc.) to varying degrees depending on the conditions of production. The high surface to
volume ratio characteristic of CB structure leads to their special surface chemistry [19]. Carbon
black is used principally for the reinforcement of rubber, as a black pigment in plastics, and for
electrical conductivity.
Graphite is an electrical conductor and also is considered the highest grade of coal,
although it is not normally used as fuel because it is hard to ignite. Graphite holds the distinction
of being the most stable form of carbon under standard conditions. Therefore, it is used in
thermochemistry as the standard state for defining the heat of formation of carbon compounds
[20]. Kwon et al. [21] investigated oxygen- and hydrogen-containing functional groups on air-
cleaved highly oriented pyrolytic graphite (HOPG) and plasma-oxidized HOPG. These groups
almost completely suppress propane adsorption at 90 K. These groups can be removed by
9
-
7/29/2019 MS_Thesis_HLi_4_4_2008
22/100
thermal treatment (500 K) and lead to more than an order of magnitude increase in adsorption
capacity. Morphological heterogeneity of plasma-oxidized HOPG provides larger surface area
for adsorption as well as higher energy binding sites [21].
Activated carbons are excellent adsorbents with their specific application related to
the properties of molecules to be removed/adsorbed. Microporous carbons favor the
sorption/separation of light gases, while carbons with broad pore size distributions favor the
removal of toxins or other large organic molecules. During the application of activated carbon,
other features, such as surface chemistry, should also be taken into account [22].
Carbonaceous materials are commonly obtained from various organic precursors
including peat, wood, coconut shells, polymers, etc.. During the process of physical activation,
the precursor is carbonized at the temperature ranging 500 700 C and then activated with
steam or carbon dioxide at the temperature ranging 700 900 C. During the process of chemical
activation, different chemicals including phosphoric acid, zinc chloride or potassium hydroxide
are mixed with the precursor and then carbonized under various temperatures. The different
processing gives rise to many types of carbons with specific applications. Carbons are
impregnated with chemicals such as KI, KMnO4, KOH or NaOH to improve the efficiency of
sorbents. Different manufacturing process produces activated carbons with different features,
including surface area, porosity, surface chemistry, etc. [23].
The wide application of activated carbon is due to their large surface area (around
1000 m2/ g) that supports many different chemistries, high pore volume, high density of carbon
atoms in graphite-like layers, and their catalytic impact in various chemical reactions. The
aforemention reasons that activated carbons could lead to fast and complete dissociation of
hydrogen sulfide in basic environments and lead to its oxidation to elemental sulfur. A
10
-
7/29/2019 MS_Thesis_HLi_4_4_2008
23/100
combination of porosity and surface chemistry leads to the oxidation of hydrogen sulfide mainly
to sulfuric acid which can be removed by water washing [23].
Bandosz et al. [24] used original activated carbons which were washed in a Soxhlet
apparatus with distilled water to remove soluble components before sulfurization experiment to
remove H2S. Results of those experiments showed that surface functional groups containing
oxygen, phosphorus and porosity of carbon significantly contribute to the process of H2S remove.
It was also found that functional groups present on the carbon surface act as catalyst for the
oxidation of H2S. Contact of physically adsorbed H2S with adsorbed water probably results in
oxidation of sulfur and formation of soluble sulfur species. The results show that neutral carbons,
with developed microporosity which may enhance the physical sorption process performed
worse as H2S adsorbents on the basis of either volume or weight compared to the original
carbons.
In general, activated carbons have micropores, mesopores and macropores hence the
adsorption rate is limited by high diffusion resistance. While activated carbon fibers typically
have micropores on the surface about 15 m diameter fiber, the adsorption rate of activated
carbon fibers is mostly controlled by rapid intraparticle diffusion so that diffusion resistance can
be neglected [25].
Activated carbons are widely used as adsorbents of gases because of their large
surface area and a big number of functional groups on their surface. When molecules are
adsorbed on the carbon surface, they are bound by Van der Waals forces that work cooperatively
in small pores. Moreover, the surface of activated carbons promotes catalytic/oxidative reactions
due to the basic and acid functional groups thus aiding the adsorption and catalysis for H2S
oxidation [26].
11
-
7/29/2019 MS_Thesis_HLi_4_4_2008
24/100
ACFs (Figure 3) have special advantages compared to the usual granular activated
carbon: (i) narrow pore size distribution, (ii) larger surface area, (iii) faster adsorption kinetics. In
addition, their sulfur species adsorption capacity can be increased by the formation of suitable
nitrogen-based surface groups via reaction with ammonia gas at high temperature [8]. Copper
compounds have been found to aid ACF in increasing the adsorption capacity toward H2S and
COS [25].
Figure 3. Scanning electron microscopy picture of activated carbon fibers used in this research.
ACFs consist of long thin sheets of graphite-like carbon. A single ACF filament is a
thin tube with a diameter of 58 micrometers and consists almost exclusively of carbon.
Several different chemicals have been used to impregnate sorbents to facilitate H2S
adsorption and catalytic breakdown including NaOH, KOH, Na2CO3, K2CO3 and KI [26]. It has
12
-
7/29/2019 MS_Thesis_HLi_4_4_2008
25/100
been observed that the H2S breakthrough capacity increases 4-5 times with increasing loading of
NaOH until maximum capacity is reached at about 10% NaOH. By increasing the pH value of
the carbon, NaOH causes HS-
ion to be oxidized to elemental sulfur or sulfuric acid, until all of
the NaOH reacts with H2SO4 [26]. However, this hydroxyl groups are completely used up and
the capacity is lost without regeneration with more OH-.
2.2.2 The Usage of Metal Oxide to Remove H2S
Hee Kwon, et al. [5] mixed zinc oxide, titanium dioxide, nickel oxide, and cobalt oxide with an
inorganic binder bentonite to fabricate Zn-Ti-based sorbents to remove H2S. These sorbents
showed excellent sulfur-removing capacities without deactivation even after 15 cycles of
sulfurization and regeneration, while the conventional Zn-Ti sorbent deactivated easily. The
nickel and cobalt in the sorbents worked as active sites during the sulfurization process, and also
stabilized the sorbent structure to prevent a change of physical properties of the sorbents during
multiple regenerations. The cobalt additive increased the regeneration capacity of the Zn-based
sorbents by supplying heat to initiate the regeneration while the nickel additive prevented SO2
slippage that originated from the cobalt sulfate.
One of the crucial factors for successful high-temperature desulfurization technology
is effective removal of H2S from fuel gases in the temperature ranging from 600 to 800 C.
Mixed metal oxides such as ZnFe2O4 can increase the H2S removal efficiency and regenerability
of the sorbent. ZnO-based sorbents could impart stability to ZnO against reduction to elemental
zinc which causes undesired structural modifications of the sorbent in highly reducing coal gas
streams. Besides ZnO-based sorbents, copper oxide that can be used at much higher temperature
than ZnO has also been investigated [27].
13
-
7/29/2019 MS_Thesis_HLi_4_4_2008
26/100
Iron-containing catalysts are promising for the selective oxidation of H2S. Fe and
Fe/Cr catalysts supported on -alumina or silica are used commercially as catalysts for the
SUPERCLAUS process. When using iron-containing catalysts, the reaction selectivity decreases
due to iron sulfide formation under the influence of the reaction medium. Through X-ray
diffraction (XRD) and X-ray photoelectron spectroscopy (XPS), it was determined that iron
disulfide FeS2 is active in the oxidation of H2S to SO2 [28].
Rare earth elements include 15 lanthanide elements (La, Ce, Pr, Nd, Pm, Sm, Eu, Gd,
Tb, Dy, Ho, Er, Tm, Yb, Lu) together with Sc and Y. The lanthanides have a wide range of
applications [29]. Lanthanum and Ce (III) oxides, have excellent sulfidation thermodynamics in
realistic reformate gas compositions, such as the ones produced by steam reforming, autothermal
reforming, or partial oxidation of heavy oils, diesel, jet fuels, or by coal gasification [30].
Li et al. [29] investigated the catalytic oxidation of hydrogen sulfide to elemental
sulfur on four rare earth orthovanadates (Ce, Y, La and Sm) and three magnesium vanadates
(MgV2O6, Mg2V2O7, and Mg3V2O8). Among the rare earth orthovanadates, the sulfur yield
decreased in the order CeVO4 > YVO4 > SmVO4 > LaVO4. It was found that sulfur yields of the
rare earth orthovanadates and MgV2O6 were superior to those of vanadium oxide.
Flytzani-Stephanopoulos et al. [30] developed reversible adsorption of H2S on cerium
and lanthanum oxide surfaces over many cycles at temperatures up to 800 C at very high space
velocities to avoid bulk regeneration of the sorbent with its structural complexities. The
adsorption and desorption processes are very fast, and removal of H2S to a low level (< ppm) is
achieved at very short (millisecond) contact times. Any type of sulfur-free gas, including water
vapor, can be used to regenerate the sorbent surface. The capacity of H2S uptake could reach 1.5
mg S/g Sorbent.
14
http://en.wikipedia.org/wiki/Celsiushttp://en.wikipedia.org/wiki/Celsius -
7/29/2019 MS_Thesis_HLi_4_4_2008
27/100
Table 2. Comparison of the application of metal oxides in desulfurization. Preparation methods
of metal oxide catalysts, advantages and disadvantages in desulfurization with these
catalysts are listed.
Metal Preparation Advantage Disadvantage Ref.
CaO Calcination of
CaCO3
Can remove H2S under high
temperature and pressure
Earlier transition occurs
at higher temperature
[6]
Ce Support -Al2O3,SiO2 and TiO2
impregnated in
Ce(NO3)3
Sorbent was convertedto by-product Ce2SO2.
[6]
Co Support -Al2O3,
SiO2 and TiO2impregnated in
Co(NO3)2
High calcination
temperature leads toformation of CoAlO4.
[6]
Cu Support -Al2O3,SiO2 and TiO2
impregnated in
Cu(NO3)2
Can be used 100% in alltemperature range
[6]
Cu Bentonite binders Able to migrate to carbon
micropores and activateoxygen
Oxidation of H2S to
SO2,
[31]
CuO andCu2O
Reduce H2S from several
thousand ppm to sub-ppm
Being reduced toelemental copper by the
H2 and CO contained in
fuel gasesSintering of layers and
causing structural
degradation
[27]
CuCr2O4 Chromium nitrate
mixed with acopper nitrate
solution
Good tthermodynamic
stability and slow reductionkinetics;Remove H2S to less than
5-10 ppm at 650-850 C
[27]
15
-
7/29/2019 MS_Thesis_HLi_4_4_2008
28/100
Table 2 (continued)
CuO-
CeO2
Cerium nitrate
mixed with coppernitrate solution
High H2S removal
efficiency, to less than 5-10ppm at 650-850 C
Easily reduce to CeO2-x, [27]
Fe Bentonite binders An improvements in
mechanical properties
Unable to activate
oxygen
[31]
Fe Fe(NO3)3 withsilica
Catalysts calcined at highertemperatures are more stable
against deactivation
Calcination at highertemperature reduces
catalytic activity of
catalysts in the H2Soxidation
[28]
MgV2O6,
Mg3V2O8
Mg(NO3)26H2O
and NH4VO3,
mixed with Sb2O4
Sulfur yields of magnesium
vanadates improved
significantly
[17]
Mn Support -Al2O3,
SiO2 and TiO2
impregnated in
Mn(NO3)2
Can be used up to100%
Reduction of Mn2O3 to
Mn3O4 can provide
additional energy to enhanceH2S sorption.
[6]
Zn Support -Al2O3,SiO2 and TiO2impregnated in
Zn(NO3)2
Reduced to metalliczinc by H2/CO andevaporate
[6]
Zn Bentonite binders An improvements in
mechanical properties
Not able to activate
oxygen
[31]
ZnO Zn(NO3)26H2O Muticycle tests of sulfidation
and regeneration could be
carried out
The rate of ZnO pellet
sulfidation is limited by
the internal masstransfer resistance
[32]
ZnFe2O4 Combining ZnO
and Fe2O3
Component reduction at
high temperatures
[27]
ZnTixOy Stable without reduction toelemental zinc at higher
desulfurization
[27]
16
-
7/29/2019 MS_Thesis_HLi_4_4_2008
29/100
2.2.3 The Usage of Impregnated Sorbents to Remove H2S
The amount of hydrogen sulfide adsorbed onto modified sorbents is larger than that adsorbed
onto raw sorbents. The increase in the amount adsorbed onto modified sorbents may be due to
the chemical interaction of the additives on the adsorbent and H2S [33]. Impregnated sorbents
have some disadvantages due to corrosive nature of impregnants such as KOH or NaOH [34].
Huang et al. [9] modified activated carbons via incipient wetness with copper nitrate
solution. In terms of the impregnation process, they proposed the mechanism of reactions
between impregnant solution and activated carbon shown in Figure 4. It was suggested that
positive charges are attracted by the activated carbon surface leading to the increase in pH value
of the solution. Cu(OH)2 formed due to water dissociation shifts, depositing onto the surface of
activated carbon. Cu(OH)2 deposit plays an important role in catalysis and oxidation of H2S.
17
-
7/29/2019 MS_Thesis_HLi_4_4_2008
30/100
18
-
7/29/2019 MS_Thesis_HLi_4_4_2008
31/100
Figure 4. Schematic of movement between impregnated solution and sorbents: a) Copper nitrate
dissociates to Cu2+ and NO3- ions in the solution which is used to impregnate
activated carbons; b) the negative charge on the surface of unmodified activated
carbon attracts H+
ions in the solution resulting in pH increase of the solution; c)
dissociated Cu2+
are attracted by activated carbon and combine with OH-
to form
Cu(OH)2; d) Cu(OH)2 deposit onto the external surface and inside the pores of
activated carbon [9].
Sorbents impregnated with caustic materials such as KOH and NaOH are widely used
for adsorption of H2S. The modification by impregnation of sorbents with metal hydroxide
solutions resulted in a decrease in porosity, especially in micropore volume, because of
adsorption or re-adsorption of metals in small pores [31]. Carbon support matrix actively adsorbs
H2S, which may greatly increase the H2S removal efficient of sorbent [35]. Other way to increase
H2S adsorption capacity is through impregnation of carbons with oxidants such as KI or KMnO4
which promote oxidation of H2S to sulfur. The capacity for physical sorption of carbon may be
19
-
7/29/2019 MS_Thesis_HLi_4_4_2008
32/100
significantly decreased due to additive compounds because the micropores, which contribute to
sorption efficiency are not available [24].
2.2.4 Other Methods to Remove H2S
The well-developed physicochemical techniques typically used for the removal of H2S have their
advantages and disadvantages. Other alternations including a variety of biological
desulfurization processes have been proposed including the use of phototrophic, heterotrophic
and autotropic microorganisms. Various processes have been developed and compared based on
the differences in nutritional requirements and abilities to catalyze specific reactions. Very few
biological processes are flexible and rapid enough on a large scale, and much work related to the
application of biological methods for treatment of H2S still needs to be done [36]. One such
method involving anaerobic, inorganic acid gas bioconversion though the photosynthetic van
Niel reaction is shown in Figure 5 via a flow diagram for a phototrophic microbial acid gas
bioconversion plant.
2n H2S + n CO2 )(hv 2n S + n (CH2O) + n H2O
H2S + 2 CO2 + 2 H2O )(hv 2 (CH2O) + H2SO4
20
-
7/29/2019 MS_Thesis_HLi_4_4_2008
33/100
Figure 5. Flow diagram for a phototrophic microbial acid gas bioconversion plant, where 1,
Photolithotrophic reactor; 2, dark reactor; 3, settler, 4, centrifuge; 5, anaerobic
methane generation. [36].
More rapid and non-biological technologies are also being developed. For example,
mesoporous materials can be generally used as adsorbents to remove impurities from gases.
Melo et al. [37] studied the use of Zeolite 13X and Zinox 380 to remove H 2S at 25 C. X-ray
fluorescence and atomic absorption, X-ray diffraction, particle size distribution analyses and FT
Infrared spectroscopy were employed to characterize these sorbents. The results show that both
materials have promising future in the removal of H2S from natural gas.
Demmink et al [38] studied the oxidative absorption of H2S into the solutions of ferric
chelates of ethylenediaminetetraacetic acid (EDTA) and hydroxyethylethylenediaminetriacetic
acid (HEDTA) using a new penetration model for mass transfer parallel to chemical reaction.
The absorption of H2S into aqueous solutions is related the oxidation of H2S:
21
-
7/29/2019 MS_Thesis_HLi_4_4_2008
34/100
H2S (g) H2S (aq)
H2S + 2Fe3+
Ln-
2Fe2+
Ln-
+ 2H+
+ S
where Ln-
is an organic ligand, in this case ethylenediaminetetraacetic acid (EDTA, n = 4) and
hydroxyethylethylenediaminetriacetic acid (HEDTA, n = 3). Limtrakul et al. [39] also used [Fe3+
EDTA] to remove H2S (Figure 6). The removal reaction takes place in the riser and regeneration
is performed in the downcomer. Iron (II) ethylenediaminetetraacetic acid [Fe2+
EDTA] could be
regenerated to [Fe3+
EDTA] with a conversion of up to 79%. It was found that the conversion of
H2S decreased with increasing superficial gas velocity in the riser, which is due to the decrease
in the liquid-phase circulation time. Desulfurization and regeneration can be carried out in the
same vessel, which would reduce the capital costs. H2S is injected at the bottom of the riser,
where the reaction occurs, and injecting air in the downcomer for catalyst regeneration or vice
versa. The gas holdup in the riser is more than that in the downcomer, and the resulting density
or pressure difference leads to a natural liquid circulation from the downcomer the riser without
any pump. The gas-lift reactor is economical and has no moving parts.
22
-
7/29/2019 MS_Thesis_HLi_4_4_2008
35/100
Figure 6. Gas-lift reactor where H2S is injected at the bottom of the riser and air injected
in the downcomer for catalyst regeneration. The removal reaction takes place
in the riser and regeneration is performed in the downcomer. H2S is injected at
the bottom of the riser, where the reaction occurs, and injecting air in the
downcomer for catalyst regeneration [39].
23
-
7/29/2019 MS_Thesis_HLi_4_4_2008
36/100
2.3 EFFECT OF PROCESS CONDITIONS ON H2S REMOVAL
During the process of removing H2S, there are two main kinetic steps. The first is controlled by
oxidation/adsorption reaction while the second is controlled by the by-products which reduce the
carbon site accessibility [34]. In the first step, the amount of water introduced, sulfurization
temperature, pressure of influent gas mixture, amount of catalysts used, and so forth can affect
the H2S removal efficiency. In terms of catalyst, the surface area, pore width, pore volume, and
functional groups on catalysts have certain effect on H2S adsorption/oxidation.
Bashkova et al. [40] found that the selectivity for oxidation of H2S to elemental sulfur
is dependent on the textural, structural, and chemical characteristics of the carbon catalyst.
Catalytically-active nitrogen species in the carbon postponed breakthrough time and increased
capacity for H2S uptake but inversely catalyzed the early formation of SO2 and COS into the
treated fuel gas stream. The significant micropore volume and narrow pore width of carbon aid
the capture of elemental sulfur and SO2 formed and the retaining of these oxidation products for
a longer time. Furthermore, they found that activated carbon could convert COS to sulfur due to
a large amount of highly reactive basic groups on the carbon surface.
2.3.1 Water Vapor Effects
The role of the humidity is highly beneficial for the removal of hydrogen sulfide as it removes
the by-products formed. Trace water in process gases leads to the acidification of the carbon
surface in the presence of oxygen or carbon dioxide, and hence restrains the dissociation of
hydrogen sulfide. Under humid condition, the oxygen or carbon dioxide results in a faster
24
-
7/29/2019 MS_Thesis_HLi_4_4_2008
37/100
reaction kinetic. The acidification of the carbon surface is counterbalanced by the dissolution of
the by-products [34].
The inlet gas containing humidity leads to the condensation of water in the pore
structure resulting in dissolution of O2 and H2S gases into the water film where oxidation of H2S
subsequently happens. The HS-ions and O2 diffuse inside a thin water film, which increases with
the polar groups present on the surface due to the formation of hydrogen bonds between the polar
species and adsorbed water. Single water molecules are adsorbed on the boundary surface. The
dissolution of H2S to HS-
is related to the amounts of water and the pH value of the water film.
Acidic carbon surface reduces the dissociation of H2S to HS
-
ions while basic carbon surface
promotes the dissociation of H2S to HS-ions. Basic condition results in a higher concentration of
HS-ions, which are then oxidized to sulfur polymers having a chain or ring-like structure [41].
The presence of moisture in gas stream plays an important role as a medium to
enhance oxidation of H2S. The dissolved H2S reacts with O2 and converts to sulfur species in the
water film on the surface of carbon. Under moisture atmosphere, additive metal oxides could
oxidize H2S though competition of adsorption may occur [9].
Under dry condition the adsorbed H2S is oxidized in a substitution reaction by oxygen
functional groups present on the carbon surface. Under wet condition, the reaction pathway
begins by the dissolution of H2S and oxygen in a thin water film on the adsorbent surface or in
the pores and is followed by a radical oxidation reaction. The highest adsorption capacity could
be obtained under a wet air atmosphere with the strongest basic carbon material [41].
Huang et al. [9] conducted parallel experiments of sulfurization of activated carbons
(impregnated by Cu(NO3)2) with gas mixture containing different relative humidity, including 0,
40-50% and 70-80%. They observed that the H2S uptake capacity decreased with the increase in
25
-
7/29/2019 MS_Thesis_HLi_4_4_2008
38/100
relative humidity. Such behavior is attributed to the competition of adsorption between H2S and
H2O on the surface of sorbent. Adsorption is an exothermic process which leads to the difficulty
of condensing water vapor to form water film [9].
2.3.2 Temperature and Surface Area Effect
Bukhtiyarova et al. [28] studied the effect of the calcination temperature on the iron oxide
catalysts prepared by impregnation of silica with iron (III) nitrate for H2S oxidation. It was found
that the increase in the calcination temperature decreases the catalytic activity of the Fe2O3/SiO2
catalysts in H2S oxidation, which has been ascribed to the agglomeration of iron oxide particles
and a subsequent decrease in active surface sites. However, the increase in the calcination
temperature makes the catalyst more stable towards the sulfidation of Fe2O3 to iron disulfide
phase.
ACFs are highly microporous solids with a small external surface area and very little
mesoporosity. Ishii et al. [42] found that the surface area of ACF decreases with the increase in
treatment temperature (1200 C). The surface area of ACF decreased from 1460 m2/g to 780
m2/g and treated ACF has slit-shaped micropores due to micrographitic structures. Heat
pretreatment of ACFs is believed to remove functional groups on the carbon surface and create
additional surface area, which is helpful to adsorption [43]. However, increasing the treatment
temperature decreases the adsorption capacity of ACF due to the collapse of micropores [42].
The high temperature treatment could partially gasify the micropore walls and thus increase the
micropore width. Nevertheless, some micropores become closed pores with reconstruction of
micrographitic structure in crystal growth leading to micropore volume decreased [42].
26
-
7/29/2019 MS_Thesis_HLi_4_4_2008
39/100
Chauk et al. [1] used CaO, which was obtained from calcination of CaCO3, to remove
H2S at the temperature ranging from 650 C to 900 C. It was found that sulfidation conversion
is very sensitive to reaction temperature and the rate of sulfidation at the beginning increased
significantly with temperatures. At higher temperature, the tapering off of the overall reaction
rate is more drastic than that at relatively lower temperatures. In the beginning, the surface
reaction may control the overall rate of CaO sulfidation. The rate of surface reaction increases
exponentially with temperature so that the nonporous CaS product layer builds up and occupies
pores leading to a decrease in the porosity of CaO. Therefore, diffusion resistance increases and
product layer diffusion begins to control the overall rate of H2S uptake.
Cal et al. [35] studied the effect of temperature on H2S adsorption at lower
temperature (400-600 C) with Zn-impregnated carbon sorbents. They found that the temperature
difference in this range had no effect on breakthrough time, which was ascribed to principles of
chemical adsorption that required a certain amount of activation energy. Physical adsorption of
H2S remains constant at lower temperatures (< 550 C).
Desulfurization at much lower temperature (23-25 C) was studied using
inexpensive-waste-based sorbent such as wood or coal fly ash. For example, elemental sulfur
was the end product of H2S oxidation but the catalytic decay occurred due to surface pores being
occupied by elemental sulfur and associated reduction in surface area (44.9-1.4 m2/g) of fly ash.
Catalyst regeneration with hot water (85 C) washing was elevated, but only half of the original
H2S oxidation activity could be regenerated [44].
27
-
7/29/2019 MS_Thesis_HLi_4_4_2008
40/100
2.3.3 Impact of Surface Functional Groups
Several methods have been used to detect and quantify oxygen functional groups on carbon
surfaces, such as Infrared spectroscopy (IR), X-ray photoelectron spectroscopy (XPS), thermal
desorption spectroscopy (TPD), elemental analysis and Boehm titration [45]. Oxygen-containing
sites exist in the form of organic functional groups, which can be acidic, basic, or neutral, and
could be catalytic centers for H2S oxidation. The H2S removal capacities are enhanced by the
presence of these oxygen functional groups [41].
Feng et al. [45] used covalent fluorescent labeling of surface species (FLOSS) to
detect relatively low amounts of surface functional groups (OH, COOH and CHO) on activated
carbon fiber surfaces. The chromophores were attached to the surface through reaction with each
type of surface functional group. FLOSS indicated the presence of COOH and CHO groups on
the ACF fiber surface while the infrared spectrum and the X-ray photoelectron spectrum could
not detect the existence of such few groups. However FLOSS could only detect exposed
functional groups and not functional groups inside smaller pores.
Spatial location, strength of interaction, and availability of surface oxygen groups are
important parameters in determining the dynamic adsorption performance of the carbons. The
total acidity, including carboxylic, lactonic and phenolic groups present on each carbon surface,
increases after oxidation and it leads to the decrease of H2S capture [46].
All oxygen groups on the carbon surface act as catalysts for the oxidation reaction,
but the basic groups play a more significant role in the process of H2S adsorption because basic
surface ensures a better immobilization of H2S molecules. H2S is an acidic gas and the presence
of basic groups contributes to the immobilization process. Thus the local pH in the pore structure
28
-
7/29/2019 MS_Thesis_HLi_4_4_2008
41/100
affects H2S adsorption/oxidation and the basic oxygen functional group concentration is an
important parameter for H2S uptake [41].
2.4 MECHANISM OF H2S REMOVAL BY CATALYTIC REACTIONS
The Claus reaction has been the standard method used in industry to remove H 2S from
waste/process gas streams. It is well known that the following basic reactions occur in the Claus
process [6]:
2 H2S + O2 (2/n) Sn + 2 H2O
(1/n) Sn + O2 SO2
2 H2S + 3O2 2 SO2 + 2 H2O
2 H2S + SO2 (3/n) Sn + 2 H2O
Ammonia which is a basic compound is also used to react with H2S to form ammonium sulfide
[(NH4)2S] which can be oxidized to produce elemental sulfur [6]:
H2S + 2 NH3 (NH4)2S
(NH4)2S + (1/2) O2 S + 2 NH3 + H2O
When NaOH [26] is used to impregnate activated carbons, NaOH shifts the
dissociation of hydrogen sulfide forward, thus increasing concentration of HS-
ions where
Ka1=9.6 x 10
-8
and Ka2= 13 x 10
-14
, shown as below
29
-
7/29/2019 MS_Thesis_HLi_4_4_2008
42/100
HS-ion is promotes oxidation of hydrogen sulfide to elemental sulfur, SO2 or sulfuric acid in the
presence of water. In addition, elemental sulfur, Sx, or polysulfides, HSxSH, deposited on carbon
might be active sites for H2S oxidation and catalyze the process.
On the surface of NaOH-impregnated carbon
Cal et al. [35] suggested that there are mainly three possible reactions for H2S uptake
by carbon-based metal-containing sorbents (involving carbon support, oxygen-containing active
sites and additive metal), as shown below:
1) Carbon: adsorption reaction between sulfur and carbon active sites, which could be identifiedby oxygen desorption.
C + H2S C-S + H2
30
-
7/29/2019 MS_Thesis_HLi_4_4_2008
43/100
2) Carbonoxygen sites: a substitution reaction in which sulfur replaces oxygen in the carbonoxygen bond. Higher surface-oxygen density leads to greater possibility of this substitution
reaction.
C-O + H2S C-S + H2O
3) Metal: reaction between sulfur and active metal because H2S is very reactive with somemetals.
C-M + H2S C-M-S + H2
The proposed mechanism of the adsorption of H2S on Pt anodes proceeds through
instant dissociation of H2S on the Pt to produce PtS species [47].
H2S + Pt Pt-S + H2
The amount of sulfate formed is so small that it could be negligible for ordinary process.
However, for fuel cells they will eventually kill the catalysts and render them useless.
H2S + 4 H2O SO42
+ 10 H+
+ 8 e
The oxidation charge of the adsorbed Pt-S is assumed to correspond to the total oxidation of
sulfur species and 6e
are used in the reactions below.
Pt-S + 3 H2O SO3 + 6 H+
+ 6 e + Pt
Pt-S + 4 H2O SO42 + 8 H
++ 6 e + Pt
31
http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml3&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=07dcf3de3712635dbd6bf13fe52a955dhttp://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1http://www.sciencedirect.com/science?_ob=MathURL&_method=retrieve&_udi=B6TF6-4DKM01W-1&_mathId=mml2&_user=88470&_rdoc=1&_acct=C000006998&_version=1&_userid=88470&md5=3c907a2ac4b01397a55fd1bb33776be1 -
7/29/2019 MS_Thesis_HLi_4_4_2008
44/100
Sulfurization of porous ZnO is usually performed at temperatures ranging from 150 to
450 C. Sulfurization kinetics and the capacity of H2S removal increases with an increase in
temperature and it reaches maximum at 450 C, followed by a reduction as the temperature
continues to increase [48]. Sulfurization of ZnO by H2S can be described by the steps shown in
Figure 7:
Figure 7. Sulfurization of ZnO with H2S, including dissociation of H2S, sulfurization of ZnO
and neutralization of yielded OH-
from sulfurization and adsorbed H+
form H2O (as
explained below) [48].
Step 1: H2S molecule adsorbs and dissociates at the active sites on the solid surface. Depending
on the surface structure and degree of sulfidation the active sites can be ZnO or ZnS.
H2S + 2 H+ + HS
-
Step 2: Reaction between adsorbed HS-and ZnO.
HS- + ZnO = ZnS + OH
-
Step 3: Yielded OH-and adsorbed H
+form H2O and active sites are reclaimed for the next cycle
after water desorption.
OH- + H
+ = H2O + 2
32
-
7/29/2019 MS_Thesis_HLi_4_4_2008
45/100
Ion migration was used to explain the sulfidation mechanism, as shown in Figure 8:
Figure 8. Ion migration and sulfidation penetration. HS-
ion penetrates the ZnS formed as
external layer leading to the internal ZnO being sulfurized to ZnS [48].
Step 1: H2S adsorbs on two active sites of ZnS on a completely sulfided grain surface.
H2S + 2 H+ + HS
-
Step 2a: H+
reacts with ZnS and forms HS-.
H+ (ZnS) = Zn2+ + HS-
The formed HS-migrates to the boundary of ZnS and reacts with ZnO.
HS-+ ZnO = ZnS + OH
-
The formed OH-migrates and combines with the adsorbed H
+formed in Step 2b.
Step 2b: The adsorbed HS-on the surface combines with Zn
2+left by proton dissociated.
HS- + Zn
2+ = ZnS + H
+ +
The sulfidation of the second layer of ZnO is in process.
Step 3, The formed OH-
in Step 2a migrates and combines with the adsorbed H+
formed in Step
2b and active sites are liberated through water desorption.
OH- + H
+ = H2O + 2
33
-
7/29/2019 MS_Thesis_HLi_4_4_2008
46/100
2.5 SUMMARY
Catalytic desulfurization is essential for advanced industrial processes to fully take advantage of
all outputs from processing fuel feedstocks as well as to prevent deleterious consequences of
nonaction. Technology currently used to adsorb and oxidize hydrogen sulfide in these gas
streams include using activated carbon as a highly effective sorbent and as a support for
desulfurization catalyst metals including copper, iron, zinc, and palladium. ACFs have
significant capacity for hydrogen sulfide uptake but are not used as extensively as activated
carbons. It has been shown that hydrogen sulfide adsorbs onto the ACF surface and is oxidized
to elemental sulfur or sulfur compounds, including COS, SO2, CS2 or SO42-
. To the best of our
knowledge, the mechanism and role of active surface chemistries on ACF surfaces in the
elimination of H2S is limited.
To clarify the quantity and nature of the functionalities on carbon surface is helpful to
understand their effect on sulfurization. Boehm Titration is frequently used to investigate
functionalities involving acid functional groups (CHO, COOH) and basic functional groups, as
many oxygen functional groups exist in tremendously varied quantities on the surface of
different carbons. Further, determining and quantifying carbon surface functional groups aids in
interpreting the reaction between hydrogen sulfide and carbon surface. The effect of temperature
on H2S uptake by carbons is critical in the application of syngas clean-up.
Lanthanum and Ce (III) oxides have excellent sulfidation thermodynamics in realistic
reformate gas compositions, especially the ones produced by steam reforming, autothermal
reforming, or by coal gasification. Although these rare earth metal oxides have potential to
remove sulfur and are able to be regenerated, the capacity of sulfur uptake is still far behind the
other well developed metal sorbents such ZnO, CuO-Cr2O3, and others. In terms of regeneration,
34
-
7/29/2019 MS_Thesis_HLi_4_4_2008
47/100
it is not practical to regenerate spent sorbents by taking sorbents out of the reactor, dissolving
formed elemental sulfur with carbon disulfide (CS2) solvent, washing away formed sulfate on the
sorbents with water and finally putting the sorbents back to the reactor after being dried.
In this work we use metal and impurity free activated carbon fibers (ACFs) as an
evaluation platform to determine the most important parameters for optimal desulfurization of
syngases. By changing the surface chemistry of the ACFs, the species responsible for sorption
and catalysis can be unambiguously determined. We compare the functional groups (especially
acidic and basic) typically observed on carbon black, graphite and ACFs to their role in
desulfurizing simulated syngas over the temperature range of 110 to 170 C. To develop and
improve the potential of H2S removal by rare earth metal, in this study, cerium and lanthanum
oxides were used to remove H2S from simulated syngas respectively and lanthanum oxide as
well as lanthanum chloride were combined with activated carbon fibers as sorbents to
adsorb/oxidize H2S.
35
-
7/29/2019 MS_Thesis_HLi_4_4_2008
48/100
3.0 MATERIALS AND METHODS
3.1 MATERIALS
3.1.1 Materials
Carbon black Black Pearls 460 (BP460), Black Pearls 800 (BP800, Cabot Corporation,
Massachusetts), Graphite (Alfa Aesar, Massachusetts) and activated carbon fibers, including
ACF7, ACF10, and ACF15 (International Interchange Ltd., Michigan) were used in our research.
Black Pearls and graphite were used as received, whereas activated carbon fibers were ground to
powders and then passed through a 80 mesh sieve with a 177 m opening. Cerium (IV) oxide
(Sigma-Aldrich, St. Louis, Missouri), lanthanum oxide (Sigma-Aldrich, St. Louis, Missouri),
lanthanum chloride LaCl36H2O (Fisher, Fair Lawn, New Jersey) and sodium hydroxide (Fisher,
Fair Lawn, New Jersey) were used to impregnated activated carbon fibers ACF10 (International
Interchange Ltd., Michigan). The average particle sizes of La2O3 and CeO2 as measured by laser
particle size analysis Microtrac S3500 (Microtrac Inc., Montgomeryville, Pennsylvania) are
~3.04 (Figure 9) and ~2.41 m (Figure 28).
36
http://www.city-data.com/zips/01821.htmlhttp://www.city-data.com/zips/01821.htmlhttp://www.city-data.com/zips/01821.htmlhttp://www.city-data.com/zips/01821.html -
7/29/2019 MS_Thesis_HLi_4_4_2008
49/100
Figure 9. Particle size and distribution of La2O3 (2.1-68 m), with an average particle
diameter about 3.04 m.
3.1.2 Simulated Syngas
Simulated syngas was formulated by mixing ultra high pure (UHP) grade H2, N2, CO, CO2, O2
and 5% H2S (balance Ar). The syngas was based on the composition of syngas used at NETL
(Table 2) and was believed to be representative of syngas for future power generation plants, as
shown in Table 3.
37
-
7/29/2019 MS_Thesis_HLi_4_4_2008
50/100
Table 3. Composition of simulated syngas under dry and wet condition in this research,
according to the syngas composition in NETL [16].
Gas H2 O2 N2 CO CO2 H2S Ar H2O Total
Dry, % 24.4 0.4 17.4 34.6 15.2 0.4 7.6 0 100
Wet, % 24.4 0.4 16.8 34.6 15.2 0.4 7.6 0.6 100
3.2 EXPERIMENTAL SETUP AND SAMPLE PREPARATION
3.2.1 Experimental Setup
The catalytic material was placed into a fixed bed with a depth of 2 cm. This was done to
maintain a constant space velocity. Typical amounts of material used varied based on the density
of the material, ranging from 2.0 g for ACF to 12 g for graphite. The fixed-bed was supported by
a coarse size frit at the bottom of the reactor. The reactor was positioned in the middle of a tube
furnace (Model 55035, Lindberg, Riverside, Michigan). Mass flow controllers (Aalborg and
MKS, Orangeburg, New York and Andover, Massachusetts) were utilized to adjust the flow rate
of each gas such that the total flow rate was 0.5 L/min and the space velocity was 3800 hr-1
. The
initial concentration of H2S in the simulated syngas was 4000 ppm. To achieve 0.6% wet gas
conditions, a portion of the nitrogen flow was directed to the water vapor generator, as shown in
Figure 10. The sulfur content of the tail-gas was measured continuously by a residual gas
analyzer (RGA, QMS300, Stanford Research Systems), through capillary column attached to a
port on the outlet of the reactor. RGA calibration for H2S was performed prior to and after
syngas application.
38
-
7/29/2019 MS_Thesis_HLi_4_4_2008
51/100
Figure 10. Schematic of carbon sorbents breakthrough capacity test. Syngas was simulated bymixing ultra high pure (UHP) gases. When water vapor was needed, N2 was
switched to water vapor generator to bring water vapor into gas mixture. Mass flow
controllers (MFC) were used to adjust flow rate and making a total flow rate 0.5
L/min. For sulfurization, gas mixture was introduced into the fixed-bed reactor
where catalyst was positioned and the tail gases through the catalyst were
continuously monitored by the residual gas analyzer whose response time to the
change of gas composition is less than 0.5 s. For regeneration of catalyst, N 2 was
introduced into the reactor and remove absorbed sulfur species and the tail gases
were also analyzed continuously by the residual gas analyzer.
39
-
7/29/2019 MS_Thesis_HLi_4_4_2008
52/100
3.2.2 Impregnation of Activated Carbon Fibers with Lanthanum Compounds
Wet impregnation was used as the main method to modify ACF sorbents according to the
following procedures: 1) ACF-LaCl3 was prepared by mixing 2.0 g ACF10 with 100 mL
aqueous solution containing 2.0 g LaCl3 for 4 hours and then dried in an oven at 140 C
overnight; 2) ACF-La2O3-NaOH was prepared by dissolving 0.31 mg La2O3 in 5 mL NaOH
solution (1 M) in a sonicated solution and diluting with 100 mL DI water and subsequently
mixing with 2.0 g of ACF. This solution was dried in an oven at 140 C overnight; 3) ACF-
LaCl3-NaOH was prepared by mixing 2.0 g of ACF in LaCl3 solution and titrating 102 mL of
NaOH solution (0.25 M). This mixture was stirred for 4 hours by a stirrer and afterwards it was
dried in an oven at 140 C overnight.
3.3 SURFACE FUNCTIONALITIES ON CARBONACEOUS MATERIALS
Boehm titration is widely used to quantify the surface functional groups on carbonaceous
materials [24]. Acidic sites were quantified under the assumption that NaOH neutralizes
carboxyl, phenolic and lactonic groups; Na2CO3 carboxyl and lactonic; and NaHCO3 only
carboxyl groups.
Boehm titration was conducted according to the methods described by Boehm [49-
51]. Briefly, standard acidic and basic solutions: 0.05N NaHCO3, Na2CO3, HCl, NaOH, and
0.25N NaOH were prepared and HCl solution was used to standardize the other solutions via
back titration. 2.0 g of each carbon sample was mixed with 100 mL of basic or acid solution and
shaken for 24 hours. After that, carbon samples were separated by filtration and the remaining
40
-
7/29/2019 MS_Thesis_HLi_4_4_2008
53/100
basicity was determined by titration with 0.05N HCl while the remaining acidity was determined
by titration with 0.05N NaOH.
3.4 SULFURIZATION AND REGENERATION EXPERIMENT
Simulated syngas was dosed to 2.0 g of each carbon sorbent in a down-flow fixed bed quartz
reactor at temperatures ranging from 25 to 210C. The inlet concentration of H2S was tested by
RGA before sorbent sulfurization by sending the simulated syngas directly to the capillary via a
bypass channel. For sulfurization, the gas mixture was switched to the fixed bed reactor and the
effluent H2S concentration was monitored with time. The experiment was shifted to the
regeneration cycle when the outlet concentration of H2S was equal to the inlet concentration, by
venting the syngas and exposing the carbon to N2. In this fashion we characterized desorption of
sulfur species, regeneration, and subsequent sulfurization and regeneration
Regeneration temperature was 400 C after sulfurization ranging at the temperature of
110 C or 170 C. When total breakthrough was reached, influent gas was switched from
simulated syngas to N2 only and temperature increased to 400 C for the regeneration cycle.
After regeneration was finished, temperature was decreased to sulfurization temperature and
syngas was introduced again. Each sorbent was treated for three cycles.
41
-
7/29/2019 MS_Thesis_HLi_4_4_2008
54/100
3.5 THERMOGRAVIMETRIC ANALYSIS
Thermal gravimetric analysis (TGA 7 - Perkin Elmer, Waltham, Massachusetts) was performed
on select sulfurized carbon samples by heating from room temperature to 600 C at different
scan rate (2, 5, 10, 20 or 40C/min) . Based on the Arrhenius equation Ea = - R T ln (k/A),
plot